{"id":"34bf1237-3701-498b-b36a-b278ec69f1b7","arxiv_id":"2505.22624","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"First EIS spectra of supra-arcade downflows show they are downflowing, turbulent, 12 to 13 MK structures, with some 'stealth' downflows showing no intensity change.","lead":"Using Hinode EIS spectra of an M-class solar flare, the authors captured the first spectroscopic measurements of supra-arcade downflows since 2003. They find the dark downflowing structures are red-shifted, turbulent, and about 12 to 13 million Kelvin, and they report a new class of 'stealth' downflows with no intensity drop.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"SAD red-shift signal may be a selection effect from the fan's north-south velocity gradient; no local background subtraction is shown.","rationale":"The paper's central claim is that SADs themselves are downflowing, based on cospatial red-shift and intensity-drop signatures. The most load-bearing assumption is that the red-shift is not simply the ambient fan's velocity pattern at those locations. The reader flagged fan/scattered-light contamination; my concern is a more specific variant: the SADs are not randomly distributed in the fan, and the fan has a strong velocity gradient (north red, south blue). Without a local background subtraction, the global histogram in Fig. 3D cannot separate SAD velocity from the fan's spatial velocity structure. This is a clean, testable issue: if the red shift persists relative to a local baseline, the claim stands; if not, the evidence collapses. I therefore retain the CONDITIONAL verdict, since the concern is addressable and does not require rejecting the paper outright.","tokens_in":15079,"tokens_out":10546,"duration_ms":129554,"concrete_test":"For each SAD contour in Fig. 2, subtract the median Doppler velocity of non-SAD pixels at the same Y-position in a ±3 timestep window around the SAD passage; test whether the residual is significantly positive. Repeat for the 'stealth SAD' candidates. If the residual is not significant in either case, the claimed SAD red shift is a selection effect.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3.2 reports that the fan is predominantly blueshifted except in the north, where the SADs are concentrated; the SAD-vs-non-SAD comparison in Fig. 3D is global, not local. If the SAD detection preferentially selects the northern, intrinsically redshifted part of the fan, then the measured 2.1-8.7 km/s LOS redshifts may reflect the fan's velocity structure rather than the SADs' own motion. The 'stealth SADs' are defined purely by Doppler features without intensity drops, so they are even more sensitive to this issue. The point-to-point 1-1.5 km/s errors and 5 km/s absolute calibration make the small shifts non-negligible. A local background comparison is needed.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents new spectroscopic observations of supra-arcade downflows (SADs) in an M3.9 flare on 2022 April 2, using Hinode/EIS sit-and-stare observations of Fe XXIV 192.02 Å, Fe XXIV 255.11 Å, and Fe XXIII 263.41 Å, together with SDO/AIA imaging. The authors identify SADs as transient intensity dips, measure their Doppler velocities, non-thermal velocities, and electron temperatures, and combine LOS Doppler velocities with POS velocities from AIA to derive total velocities for four SADs. They report that SADs are redshifted relative to the surrounding fan, show enhanced non-thermal velocities, have temperatures near the fan value, and that some SAD-like Doppler features occur without intensity drops, which they term 'stealth SADs'. They also interpret a north-south Doppler asymmetry as evidence for divergence of SADs above the flare loop arcade.","tokens_in":15218,"tokens_out":9534,"duration_ms":111208,"significance":"The dataset is unique and valuable: it represents the first EIS spectroscopy of SADs since the SUMER era, and the multi-line diagnostics provide constraints that imaging alone cannot. The paper is careful in quoting uncertainties (point-to-point 1–1.5 km/s, absolute 5 km/s) and in discussing the CHIANTI-based line blending. The main claims—that the SADs themselves are downflowing, that they show enhanced turbulence, and that a subset shows no intensity drop—are interesting and would, if robust, advance the debate on the nature of SADs. However, the central Doppler-velocity result is currently not demonstrated at the required level.","major_comments":[{"comment":"The comparison of Doppler velocities between SAD and non-SAD pixels uses all non-SAD pixels in the field of view, rather than a local background at the same slit position. The paper itself states that the northern half of the Doppler map is predominantly redshifted, and the SADs are concentrated in that northern region (top half of the map). Consequently, the measured 2.1–8.7 km/s redshifts may simply reflect the ambient fan velocity gradient rather than a flow intrinsic to the SADs. This is particularly important because these shifts are comparable to the 1–1.5 km/s point-to-point uncertainty and below the 5 km/s absolute calibration. The same issue affects the non-thermal velocity histogram (Figure 3E) and the divergence interpretation (Section 5, item 2). To support the claim that the SADs are themselves redshifted, the authors must show a local control: either compare SAD pixels with non-SAD pixels at the same Y-coordinate and at the same or immediately adjacent times, or subtract a spatially and temporally smoothed background velocity from the Doppler map and demonstrate that the SAD signatures remain.","section":"Section 3.2, Figures 3D-3F"},{"comment":"The existence of 'stealth SADs' is not established. The paper identifies red-shift features that do not coincide with intensity drops as stealth SADs, using only the Doppler signature. However, the same map shows redshifted regions that are not associated with any detected SAD, and the paper has not shown that these features have independent SAD-like properties (such as coherent motion in time-distance plots or enhanced non-thermal velocities) or that they are distinct from the ambient fan's velocity structure. Without such a control, the interpretation of these features as downflowing SADs is speculative.","section":"Section 3.2 and Section 5, item 6"}],"minor_comments":[{"comment":"The wavelength of the Fe XXIV line is given as 192.02 Å in the abstract and Figure 2, but as 192.04 Å in Section 3.2; please make this consistent.","section":"Section 3.2"},{"comment":"The text says 'Figure 3D shows a map of non-thermal velocity', but Figure 3D is a histogram; the map of non-thermal velocity is in Figure 2D. Please correct the reference.","section":"Section 3.3"},{"comment":"Near Figure 3, the text mentions 'SADs i-v', but only four SADs (i–iv) are labeled; please correct the numbering.","section":"Section 3.2"},{"comment":"The column header for non-thermal velocity appears as 'V N T'; please use a clearer notation such as 'V_NT'.","section":"Table 1"},{"comment":"The line ratio is written as 'Fe XXIII 263.41 Å / Fe XXIV 255.11 Å' in some places and as 'Fe XXIV 255.11 / Fe XXIII 263.41' in others; please be consistent.","section":"Section 3.4"},{"comment":"The definition of the line-of-sight angle is given as θLOS = tan(Vlos/Vpos), which should presumably be θLOS = atan(Vlos/Vpos); please correct.","section":"Section 4"}],"recommendation":"major_revision","confidential_remarks":"The paper is a strong candidate for a letter if the Doppler analysis is tightened. The data are valuable and the authors have a good handle on uncertainties. The main concern is the lack of local background subtraction in the velocity analysis; this is addressable in revision. I would not require new observations, but the authors need to redo the comparison using local controls and temper the stealth-SAD claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Ryan,\n\nGood to have this one in hand. The paper is the first Hinode EIS spectroscopy of supra-arcade downflows, and the first spectral measurements of SADs since the SUMER papers in 2003. That alone makes it worth reading. The authors extract Fe XXIV Doppler velocities, non-thermal widths, and Fe XXIV/Fe XXIII line-ratio temperatures from a sit-and-stare slit that catches four clear SADs, and they combine the LOS velocities with AIA plane-of-sky tracking to get 3D flow vectors. They also identify a new class of 'stealth SADs'—redshifted flows with no intensity drop. The observational work is mostly careful: they flag the 5 km/s absolute wavelength uncertainty, the Fe XI blend in Fe XXIV 192, the need to bin the weak 255/263 spectra, and the ionization-equilibrium assumption. The temperature result, 12.2–13.4 MK, is consistent with earlier DEM estimates and is a genuinely new spectral measurement.\n\nThe main soft spot is exactly what the stress-test note says. The SAD-vs-non-SAD histogram in Fig. 3D is global, not local. The Doppler map shows a clear north-south gradient, with the fan itself redshifted in the north, where the SAD detections are concentrated. The measured SAD redshifts are 2.1–8.7 km/s, which is the same order as the fan's spatial gradient. Without a local background subtraction—comparing each SAD to the immediate surrounding fan at the same y and time—you can't tell whether the redshift belongs to the SAD or to the underlying fan. The stealth SAD claim is even more exposed, because those features are defined purely by Doppler shifts with no intensity contrast; any redshifted patch of fan could masquerade as one. This is fixable: compute a local background map (e.g., a smoothed version of the fan excluding SAD pixels) and subtract it, and do a null test by checking whether the 'stealth' features survive when compared to the local fan velocity distribution.\n\nA minor point: the LOS velocities are small relative to the absolute calibration, and the paper acknowledges this, but it means the angle of the 3D velocity vectors is not well constrained. That does not undermine the total velocity magnitudes, which are dominated by the POS components. The citation pattern is appropriate; the authors engage directly with the Savage et al. SADL interpretation.\n\nThis deserves a serious referee. I'd send it out, and I'd ask the authors to add the local background analysis before acceptance. The dataset is valuable, and the central claim—that the dark SADs are themselves downflowing—is plausible, but it is not yet closed.\n\nBest,","headline":"First EIS spectra of supra-arcade downflows, but the main Doppler result needs a local background comparison before it fully lands.","tokens_in":15768,"tokens_out":4281,"would_cite":true,"duration_ms":46718,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The dark downflows seen above solar flares are themselves downflowing plasma, not static voids.","keywords":["supra-arcade downflows","solar flares","EUV spectroscopy","Hinode EIS","Doppler velocities","non-thermal velocities","electron temperature","stealth SADs"],"falsifier":"Take the binned Fe XXIV 192.02 Å spectra from the SAD contours and fit a two-component model consisting of the static, brighter fan spectrum plus a low-density void with no bulk motion. If such a model reproduces the observed centroid shifts and line profiles as well as a single red-shifted Gaussian, the claim that the SADs themselves are downflowing plasma would be falsified.","tokens_in":14891,"feed_emoji":"☀️","tokens_out":4337,"duration_ms":47787,"temperature":0.7,"pith_summary":"This paper reports the first spectroscopic observations of supra-arcade downflows (SADs) since 2003, using Hinode EIS sit-and-stare data of an M3.9 flare on April 2, 2022. It tries to establish that the dark SAD streaks are not merely static low-density voids but are themselves plasma flowing downward toward the flare arcade. The evidence is Fe XXIV Doppler red shifts that align spatially and temporally with the dark intensity drops, plus total velocities of 59–83 km/s when line-of-sight and plane-of-sky components are combined. The paper also derives electron temperatures of 12.2–13.4 MK, finds enhanced non-thermal velocities inside SADs, and reports 'stealth SADs' that produce downflow Doppler signals with no measurable intensity drop. If correct, these results connect SADs more directly to the standard flare reconnection picture and give future models new observables to reproduce.","feed_headline":"Dark solar flare streaks are true downflows, new spectra show","feed_subtitle":"Hinode EIS reveals red-shifted plasma, a possible termination shock signature, and 'stealth' downflows with no darkening.","key_machinery":"The central observational machinery is the Hinode EIS Fe XXIV 192.02 Å sit-and-stare sequence, where a slit is held fixed while SADs drift across it, giving time-resolved spectra of individual structures. Doppler velocities come from single-Gaussian centroid fits calibrated to a quiet-Sun reference line, non-thermal velocities from excess line broadening, and electron temperatures from the ratio of Fe XXIV 255.11 Å to Fe XXIII 263.41 Å lines compared with CHIANTI theoretical ratios. Plane-of-sky velocities are measured by tracking the same SADs in AIA 131 Å time-distance diagrams; combining both components gives the 3D velocity profile for four SADs.","core_discovery":"Using sit-and-stare spectra from Hinode EIS, the paper finds that the darkest, most prominent supra-arcade downflows in the April 2, 2022 flare coincide with Fe XXIV 192.02 Å Doppler red shifts of 2.1–8.7 km/s along the line of sight, while the surrounding flare fan is predominantly blue-shifted. Combining these Doppler measurements with AIA 131 Å plane-of-sky tracking yields total SAD velocities of about 59–83 km/s, similar to, though on the lower end of, previously imaged SAD speeds. The paper further reports that SADs show higher non-thermal velocities than the fan (minimum about 20 km/s, with peaks above 70 km/s), electron temperatures of 12.2–13.4 MK close to the fan temperature, a north-south Doppler pattern interpreted as SADs diverging above the flare looptop, and the detection of 'stealth SADs' that have SAD-like Doppler signatures but no corresponding intensity drop. The central conclusion is that the dark SADs themselves are downflowing plasma, not static low-density wakes behind contracting loops, although the authors note that the contracting-loop interpretation could still hold if the loops are too thin or too cool to detect in their data.","pith_inferences":["Beyond the paper: if 'stealth SADs' are common, the total number of downflow events in a flare may be substantially higher than dark-feature counts suggest, and Doppler surveys could provide a more complete census.","Beyond the paper: the observed time lag between the intensity drop and the peak non-thermal velocity, with turbulence persisting after the SAD passes, could be used to estimate dissipation or drag in the flare fan if compared with MHD turbulence decay models.","Beyond the paper: because the SAD temperatures match the fan temperature, a density-sensitive line pair in future EIS observations could directly test whether SADs are in pressure balance with their surroundings, which would help distinguish void models from contracting-flux-tube models.","Beyond the paper: if the divergence pattern is produced by a termination shock, the measured SAD velocity directions may constrain the shock height and geometry when combined with the flare's magnetic field extrapolation."],"forward_implications":["If SADs are truly downflowing plasma rather than static voids, models of supra-arcade downflows must explain both the intensity deficit and the bulk downward motion of the same structure.","The north-south pattern of Doppler shifts, interpreted as divergence above the flare looptop, supports the idea that a high-altitude termination shock decelerates reconnection outflows before they reach the arcade.","SAD temperatures close to the surrounding fan imply that the dark appearance of normal SADs is primarily a density effect, not a temperature effect.","The existence of 'stealth SADs' means imaging surveys that identify SADs only by dark features will miss a population of downflowing structures with no intensity drop.","Sit-and-stare spectroscopy with a slit crossing a flare fan can catch individual SADs, offering a practical observing strategy for future spectral studies of these transient structures."],"supporting_citations":[{"why":"First identified supra-arcade downflows and gave the initial speed range, providing the baseline comparison for the velocities measured here.","marker":"D. E. McKenzie & H. S. Hudson 1999"},{"why":"The last published SAD spectral observations, whose reported high-velocity blue wing is the direct contrast for the absence of such features in the new EIS spectra.","marker":"D. E. Innes et al. 2003b"},{"why":"Proposed the SADL interpretation that SADs are low-density wakes behind contracting loops, the main alternative interpretation this paper's red-shift result challenges.","marker":"S. L. Savage et al. 2012"},{"why":"MHD modeling of a high-altitude termination shock that decelerates reconnection outflows, used to explain the observed divergence and lower SAD speeds.","marker":"C. Shen et al. 2022"},{"why":"Previous DEM-based SAD temperature limit and finding that SADs are close in temperature to the fan, both directly compared with the new line-ratio temperatures.","marker":"W. J. Hanneman & K. K. Reeves 2014"},{"why":"Fe XXI Doppler observations of red/blue shift patterns above a flare looptop, used as the interpretive template for SAD divergence beneath a termination shock.","marker":"V. Polito et al. 2018"},{"why":"CHIANTI database source used to compute the Fe XXIV/Fe XXIII temperature-sensitive line ratios.","marker":"K. P. Dere et al. 1997"},{"why":"Updated CHIANTI spectral synthesis used to verify the Fe XXIV 192.02 Å blend contribution and theoretical line ratios.","marker":"R. P. Dufresne et al. 2024"},{"why":"Identifies the Fe XI blend in the Fe XXIV 192.02 Å line, justifying the single-Gaussian fit as the blend is negligible at flare temperatures.","marker":"P. R. Young et al. 2007"}],"fun_headline_variants":["Spectra confirm dark solar flare streaks are real downflows","Hinode spectra show dark downflows are true plasma motions","Doppler data prove dark solar streak downflows are real","New spectra reveal dark flare tendrils are downflowing plasma","Stealth downflows and red shifts: new view of dark solar streaks"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the Fe XXIV spectra extracted from the binned, low-intensity SAD pixels are dominated by the SAD plasma itself, rather than by scattered light or by the much brighter surrounding flare fan, so the measured Doppler shifts, temperatures, and line widths genuinely describe the SAD rather than the fan.","fun_headline_variants_meta":{"raw":{"variants":["Spectra confirm dark solar flare streaks are real downflows","Hinode spectra show dark downflows are true plasma motions","Doppler data prove dark solar streak downflows are real","New spectra reveal dark flare tendrils are downflowing plasma","Stealth downflows and red shifts: new view of dark solar streaks"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000776,"raw_usage":{"total_tokens":3525,"prompt_tokens":1131,"completion_tokens":2394,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":747,"completion_tokens_details":{"reasoning_tokens":2308}},"tokens_in":747,"tokens_out":2394,"duration_ms":16455,"temperature":1.0,"reasoning_tokens":2308,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T13:02:46.040919+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the binned Fe XXIV 192.02 Å spectra from the SAD contours and fit a two-component model consisting of the static, brighter fan spectrum plus a low-density void with no bulk motion. If such a model reproduces the observed centroid shifts and line profiles as well as a single red-shifted Gaussian, the claim that the SADs themselves are downflowing plasma would be falsified.","supporting_citations":[{"cited_title":"K., & Musset, S","cited_arxiv_id":null,"evidence_quote":"Fe XXI Doppler observations of red/blue shift patterns above a flare looptop, used as the interpretive template for SAD divergence beneath a termination shock."},{"cited_title":"R., Del Zanna, G., Mason, H","cited_arxiv_id":null,"evidence_quote":"Identifies the Fe XI blend in the Fe XXIV 192.02 Å line, justifying the single-Gaussian fit as the blend is negligible at flare temperatures."}],"review_version":1}